In a recent article (Feldman, Smart, & Ethington, 1999), we used Holland's (1966, 1973, 1985, 1997) theory to study the extent to which the growth or achievement of college students--in terms of their gains in selected abilities and interests--was a function of the congruence or fit between these students' personalities and their major field environments. The present analysis is an outgrowth and extension of that earlier analysis. It explores and compares the influence of personality dynamics and social-environmental forces on students' change and stability during college. Holland's (1966, 1973, 1985, 1997) theory of personality types and environments assumes that most people (including college students) can be classified as one of six personality types (Realistic, Investigative, Artistic, Social, Enterprising, Conventional) based on their distinctive patterns of abilities, attitudes, and interests. The theory further provides three general propositions pertaining to college students and their development during their collegiate experience: (a) students search for and select academic environments that match their distinctive patterns of abilities, interests, and personality profiles; (b) academic environments differentially socialize students toward the acquisition of distinctive patterns of abilities, interests, and values that reflect the abilities, interests, and values presumably reinforced and rewarded by the respective environments; and (c) student achievement is a function of the congruence or fit between their dominant personality type and the academic environment. Hol land's theory is primarily psychological in origin and nature, and most research on the validity of the three propositions has focused on the first (self-selection) and third (congruence) propositions in efforts to assist students to select "proper academic majors" in which they stand the greatest likelihood of success and to better understand differential patterns of change and stability in students' abilities and interests during their undergraduate careers. Abundant evidence generally supports the validity of each of the three propositions--especially those of self-selection and congruence (see, for example, the extensive literature reviews by Assouline & Meir, 1987; Holland, 1985, 1997; Spokane, 1985, 1996; Tranberg, Slane, & Ekeberg, 1993; Walsh & Holland, 1992); and reliance on Holland's theory as a theoretical framework for studies of college effects on students is growing (see, for example, Antony, 1998; Feldman et al., 1999; Huang and Healy, 1997; Thompson & Smart, 1999). In the present analysis we are not so much interested in offering additional evidence for any one of these three propositions as in assessing the relative merits of two of them--the second (socialization) and the third (congruence). The socialization and congruence propositions provide different--if not alternative--explanations for differential patterns of change and stability in students. The socialization proposition postulates that the key element in promoting student acquisition of one rather than another set of competencies and talents is the academic environments (i.e., departments) students enter. Here, the roles of faculty members and their collective efforts to socialize students to the prevailing norms and values of their respective academic environments are the primary components to be considered, and the personality types and associated initial abilities and interests of students are of less importance and perhaps even irrelevant. That is to say, for example, that the likelihood of students collectively developing any specific repertoire of competencies and values is singularly dependent upon their entry into an academic environment that requires, reinforces, and rewards that particular repertoire. In this respect, the socialization proposition has a decided sociological orientation because of its focus on the collective group effects of academic environments. …